Skip to main content
Back to timeline
Vercel NewsSource publication:

Parallel Decisions arrives free on AI Gateway: live-web answers on companies, products and policies, with probabilities and citations

Synopsis

Parallel's Parallel Decisions Fast model is now available on AI Gateway and free through November 23, 2026. It searches the live web to answer typed questions about companies, products, policies and other changing information, returning probabilities and cited evidence for an application to act on. Questions come in three kinds: boolean questions return the probability of yes, choice questions select from options the caller defines, and score questions assess an ordered rubric; related questions share context and retrieved evidence in one request.

Source-provided article image: Parallel Decisions now available on AI Gateway for free
Vercel

Interpretation

Parallel's Parallel Decisions Fast model is available on AI Gateway and free through November 23, 2026, called under the model name parallel/parallel-decisions-fast, with no separate Parallel account required. The model reaches AI Gateway under a single model name, so a caller needs no separate Parallel account to use it. The model searches the live web to answer questions about companies, products, policies and other changing information, returning probabilities and cited evidence for the application to act on.

Callers can organize requests by question type: boolean questions return the probability of yes, choice questions pick one of the caller's options, and score questions assess an ordered rubric. Related questions in the same request share context and retrieved evidence rather than each retrieving independently. The examples ask the same web-grounded question through the AI SDK, the TypeSafe-compatible API and the OpenAI-compatible Decisions API, then read answers and citations.

Citations include source URLs, optional titles and evidence excerpts, and apply to the whole request rather than mapping individual sources to each answer. An application can act automatically above a probability threshold and send uncertain results for review. AI Gateway tracks these requests in logs and spend budgets alongside other model requests.

Perspective

The model targets typed questions about companies, products, policies and other changing information, returning probabilities and cited evidence; the material shows three calling paths, the AI SDK, the TypeSafe-compatible API and the OpenAI-compatible Decisions API, and reports no accuracy or latency figures on any task.

Citations apply to the whole request rather than mapping to individual answers, so the caller must handle how each answer corresponds to specific sources; the material also does not say whether the probabilities are calibrated.